[CP2K-user] Is there a proper tutorial on how to use FREE_ENERGY calculations?

[Marcella Iannuzzi]

CP2K, as other codes, can run MD under different conditions, for instance 
with constraints, restraints, combining different force environments  and 
so on. 
The work to extract from the generated trajectories and related properties 
the free energies is done a posteriori. 
If you read the literature you can find it out yourself. 

To calculate solvation energies you can follow the work of Cheng et al 
(https://aip.scitation.org/doi/10.1063/1.3250438)
and following works. 

I cannot say whether you really need the multiple force environment 
features. Anyway, running simultaneously different environments it uses 
combined energies and forces and generates only one dynamics. How energies 
and forces are combined is defined by the user in the input. 

Regards
Marcella


On Wednesday, November 11, 2020 at 4:33:45 PM UTC+1 Dev Rana wrote:

> Thank you Marcella! That is indeed helpful and I will look into further. 
> However, my question was more about how it's implemented in CP2K and how 
> the theory (textbooks) can/should be properly run in CP2K's code. For 
> example, can I just turn the METHOD UI on, set some COLVAR (RMSD solvent 
> atoms), and expect it to give me Gibbs free energy of solvent?
>
> If I want to set a coupling parameter (lambda), I can only set this in 
> FORCE_EVAL > MIXED > LINEAR > LAMBDA. I can also do this CDFT, but that's 
> not really what I want. Now if I set LAMBDA in LINEAR to 0.1 or any 
> d(lambda), do I run a new AIMD using that lambda? I don't really understand 
> the use of 2 FORCE_EVAL sections, and how CP2K does these calculations. Are 
> there two sets of output/trajectory files? Or do the two FORCE_EVAL 
> sections correspond to State A and State B within the definition of free 
> energy calculations?
>
> The two examples on the website are really just an introduction to using 
> collective variables and metadynamics, not really for obtaining free 
> energies. (
> https://www.cp2k.org/exercises:2015_cecam_tutorial:mtd1?s[]=free&s[]=energy 
> and https://www.cp2k.org/howto:biochem_qmmm?s[]=free&s[]=energy)
>
> Sorry, I don't mean to bombard you with questions, however, I'm trying to 
> learn how to implement the different techniques within CP2K and I'm 
> struggling with it. It's a lot of guess and test and someone who has 
> already done this can speed up my (and others who want to use this) 
> progress. If it's been published or could even provide some technical setup 
> details that could serve as a basis for computational start point. 
>
> Thank you again. I really do appreciate your help! 
>
>
> On Wednesday, November 11, 2020 at 2:53:44 AM UTC-5, Marcella Iannuzzi 
> wrote:
>>
>> Yes many
>> an example
>> https://link.springer.com/book/10.1007/978-3-540-38448-9
>> Marcella
>>
>>
>> On Wednesday, November 11, 2020 at 6:28:08 AM UTC+1 Dev Rana wrote:
>>
>>> Hello,
>>>
>>> I'm trying to use thermodynamic integration or umbrella sampling to 
>>> obtain solvation and molecule free energies. Unfortunately, there are no 
>>> good tutorials in CP2K to do this. Are there good external tutorials or 
>>> publications that can help to navigate this?
>>>
>>> Best Regards,
>>> Devyesh Rana
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20201111/39af37e0/attachment.htm>